Secure Federated Learning Framework with Dynamic Lightweight Cipher (Fed- DLC) for Resource-Constrained Networks and Data-Sensitive Applications
Abstract
Abstract Federated learning (FL) enables collaborative model training without directly centralizing raw client data; however, transmitted model updates may still expose information or be intercepted over an untrusted communication channel. Existing protection mech-anisms, including homomorphic encryption and other cryptographic techniques, can introduce computational and communication overhead that is challenging for resource-constrained deployments. This paper presents Fed-DLC, a federated learning frame-work that integrates a dynamic lightweight cipher for protecting model-parameter exchanges between participating clients and the coordinating server. The framework is evaluated using ResNet-20, ResNet-50, and InceptionV3 on MNIST, Fashion-MNIST, EMNIST, and CIFAR-10, and its utility performance and execution time are compared with standard FL and selected secure or privacy-oriented baselines. The reported results indicate that, under the experimental configuration considered, Fed-DLC preserves task performance close to the corresponding FL implementation while introducing additional computation time. The proposed method is intended as a lightweight communication-protection layer; its cryptographic strength depends on the threat model, key-management assumptions, and resistance of the underlying cipher to cryptanalysis. Accordingly, this study explicitly distinguishes empirical utility and efficiency results from cryptographic security guarantees and identifies comprehensive…
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.